Method of forming T electrode in field effect transistor

1. Field of the Invention

The present invention relates to a method of manufacturing a semiconductor device and, more particularly, to a method of forming a T electrode in a field effect transistor.

2. Description of the Prior Art

With high integration of semiconductor devices, micropatterning of gate electrodes has been advanced. However, micropatterning of gate electrodes inevitably leads to an increase in gate resistance. As a measure against this, a method of forming a T gate electrode is employed. FIGS. 1A to 1E are sectional views, respectively, showing a semiconductor chip to explain the steps in manufacturing the T gate electrode of a conventional GaAs FET.

As shown in FIG. 1A, after an operating layer 2 is formed on a semi-insulating substrate 1, part of the surface of the operating layer 2 is selectively removed by wet etching or dry etching using a photoresist pattern as a mask, thereby forming a recess step 4. An insulating film 8 made from SiO 2 is formed by LP-CVD on the obtained structure. An antireflection film 9 is formed on the insulating film 8. According to Japanese Unexamined Patent Publication No. 6-204130, this antireflection film 9 is formed by applying a silicon-based organic material by spin coating. Then, a first positive photoresist film 5A is applied on the antireflection film 9.

As shown in FIG. 1B, the resultant structure is exposed by an aligner (not shown) using a photomask (not shown). As an exposure beam, an i-line (λ=365 nm) or a KrF excimer beam (λ=248 nm) is used. In exposure, an exposure beam passing through the photomask is incident on an exposure region L in the photoresist film 5A. The antireflection film 9 below the photoresist film 5A has a low transmittance with respect to the exposure beam, and is made thick enough to prevent transmission of the incident beam to the under-layer of the antireflection film 9, or reflection thereof to an upper layer. For this reason, a pattern is stably formed without being influenced by reflection or interference of the exposure beam coming from the underlayer. Subsequently, the resultant structure is developed by an alkali developing solution to dissolve the exposure region, forming a resist pattern having a resist opening 6A in the recess step.

As shown in FIG. 1C, the surface of the antireflection film 9 to be exposed is processed by an oxygen (O 2 ) plasma, and selectively changed into an inorganic-system film 9A to attain the same etching characteristics as those of the SiO 2 film 8. Using the first photoresist film 5A as a mask, the inorganic-system film 9A and the SiO 2 film 8 are selectively removed by dry etching with SF 6 gas, thereby forming an opening portion 6B.

As shown in FIG. 1D, after the remaining photoresist film 5A is removed by an O 2 plasma, the remaining antireflection film 9A is completely removed by dry etching using SF 6 gas. A metal film 17 is sputtered on the entire surface including the opening portion 6B to fill the gate opening portion 6B. The following steps will be explained using an example disclosed in Japanese Unexamined Patent Publication No. 63-245961. The metal film 17 is made from WSi-Ti-Pt-Au. Using a second photoresist pattern 5B as a mask, an unnecessary portion of the metal film 17 is selectively removed by ion milling.

As shown in FIG. 1E, the photoresist pattern 5B is removed to form a gate electrode 17A. An Si 3 N 4 film 10 is formed by CVD, and then dry-etched using CF 4 gas to leave it on only the side walls of the gate electrode 17A. Further, the resultant structure is wet-etched using a solution containing hydrofluoric acid to remove the SiO 2 film 8. At this time, the Si 3 N 4 film 10 serves to protect the Ti layer in the metal film 17 from being etched.

The first problem in the conventional technique is that addition of the antireflection film between the photoresist film and the SiO 2 film serving as a target etching film results in an increase in number of steps by the steps of forming, etching, and completely removing the antireflection film, compared to a case of forming no antireflection film.

The step of etching the antireflection film can be performed under the same conditions as those of the step of etching the SiO 2 film. However, since the total film thickness which should be etched is larger than that in the case of etching only the SiO 2 film, the dimensional accuracy for transferring a resist pattern to the SiO 2 film decreases.

The second problem in the conventional technique is that the step of forming and processing the Si 3 N 4 film is added because a T gate electrode formation metal must be protected by the Si 3 N 4 film in order to prevent etching of the metal in the step of completely removing the SiO 2 film upon forming the T gate electrode.

The present invention has been made in consideration of the above situation in the prior art, and has as its object to provide a semiconductor manufacturing method in which the step of manufacturing a T gate electrode can be simplified, and the productivity is greatly improved.

To achieve the above object, according to the first aspect of the present invention, there is provided a method of manufacturing a semiconductor device, comprising the steps of sequentially forming an operating layer and a light-shielding film so as to form a recess step on a semiconductor substrate, forming a first photoresist film on the light-shielding film, patterning the light-shielding film using the photoresist film as a mask to form a gate electrode formation opening portion, forming a metal film on an entire surface including the opening portion, selectively etching the metal film using, as a mask, a second photoresist film formed on the metal film, thereby forming a gate electrode having a T shape in a longitudinal section, removing the second photoresist film, and removing the light-shielding film.

According to the second aspect of the present invention, there is provided a method of manufacturing a semiconductor device, wherein the light-shielding film in the first aspect is made from one material selected from the group consisting of Si, TiN, and SiC.

According to the third aspect of the present invention, there is provided a method of manufacturing a semiconductor device, wherein the light-shielding film in the first aspect is made from amorphous carbon.

According to the fourth aspect of the present invention, there is provided a method of manufacturing a semiconductor device, wherein the light-shielding film and the second photoresist film in the third aspect are simultaneously removed.

According to the fifth aspect of the present invention, there is provided a method of manufacturing a semiconductor device, wherein the metal film in the first aspect is made from Al.

As is apparent from the respective aspects, the light-shielding film used in the present invention serves to eliminate unstableness of the pattern dimension and shape caused by reflection at the underlayer in exposing the photoresist film. At the same time, since the metal film is buried in the light-shielding film, this light-shielding film also serves as a mold used in forming an electrode. Therefore, a T gate electrode can be formed by a step simpler than that of the conventional manufacturing method.

By using an electrode formation metal film which is not etched in the step of completely removing the light-shielding film of the present invention, the step of protecting the gate electrode with an Si 3 N 4 film can be eliminated.

According to the first effect of the present invention, the number of films formed on a semiconductor substrate can be decreased from two layers, i.e., a target etching film and an antireflection film in the prior art to one layer. As a result, the step of forming a film on a semiconductor substrate, and the etching step for a gate opening can be simplified. This is because the light-shielding film made from an Si film or an α-C film formed on the semiconductor substrate serves as both the target etching film for burying a gate electrode metal film and forming a T gate electrode, and the underlying antireflection film used in exposing a photoresist film.

According to the second effect of the present invention, the gate electrode metal film need not be protected by any Si 3 N 4 film to be prevented from being etched in the step of removing the light-shielding film which has already been unnecessary in forming a T gate electrode. Consequently, the number of steps in forming a T gate electrode can be decreased. This is because, in the first embodiment, an Al film is used as a gate electrode formation metal, so that a gas which etches the Si film without etching the gate electrode can be selected. Also, in the second embodiment, O 2 gas used to remove the α-C film does not etch a conventionally used gate electrode formation metal film.

The above and many other objects, features and advantages of the present invention will become manifest to those skilled in the art upon making reference to the following detailed description and accompanying drawings in which preferred embodiments incorporating the principles of the present invention are shown by way of illustrative example.

FIGS. 1A to 1E are sectional views, respectively, showing the steps in a conventional method of manufacturing a semiconductor chip;

FIGS. 2A to 2E are sectional views, respectively, showing the steps in a method of manufacturing a semiconductor chip according to the first embodiment of the present invention; and

FIGS. 3A to 3E are sectional views, respectively, showing the steps in a method of manufacturing a semiconductor chip according to the second embodiment of the present invention.

Several preferred embodiments of the present invention will be described below with reference to the accompanying drawings. FIGS. 2A to 2E are sectional views, respectively, showing the steps in a method of manufacturing a semiconductor device according to the first embodiment of the present invention.

As shown in FIG. 2A, after an operating layer 12 made from GaAs is formed by MBE on a semiconductor substrate 11 made from GaAs, the surface of the operating layer 12 is selectively removed by wet etching or dry etching using a photoresist pattern as a mask, thereby forming a recess step 14. An Si film 13 is sputtered as a light-shielding film on the obtained structure. Then, a first positive photoresist film 15A is applied on the entire surface.

As shown in FIG. 2B, the photoresist film 15A is exposed by an aligner (not shown) using a photomask (not shown). As an exposure beam, an i-line (λ=365 nm) or a KrF excimer beam (λ=248 nm) is used. In exposure, an exposure beam passing through the photomask is incident on an exposure region L in the photoresist film 15A. Since the Si film 13 below the photoresist film 15A has a low transmittance with respect to the exposure beam, the exposure beam incident on the Si film 13 is completely absorbed to prevent the exposure beam reflected by the lower surface of the Si film 13 from striking the photoresist film 15A again. Since the Si film 13 has a low reflectivity at the interface with the photoresist film 15A, deformation of a pattern caused by reflection at the underlying step, waving of a pattern side wall due to the presence of a standing wave between incident and reflected beams, and variations in dimension caused upon changes in exposure energy by the multiple reflection effect in a resist film can be prevented. Subsequently, the resultant structure is developed by an alkali developing solution to dissolve the exposure region, forming a resist pattern having a resist opening 16 in the recess step 14.

As shown in FIG. 2C, the Si film 13 is selectively removed by dry etching using SF 6 gas and the remaining first photoresist film 15A as a mask. A resist opening pattern is transferred to form an opening portion 16B, and then the photoresist film 15A is removed by an O 2 plasma.

As shown in FIG. 2D, a metal film 17 is formed by, e.g., deposition on the entire surface including the opening portion 16B. At this time, the opening portion 16B is filled with the metal film 17. As this metal film, an Al film 17 or the like is used. Using a second photoresist pattern 15B as a mask, the Al film 17 is selectively removed by dry etching using CCl 4 gas or BCl 4 gas to form a gate electrode 17A having a substantially T shape. Thereafter, the photoresist film 15B is removed using an O 2 plasma.

As shown in FIG. 2E, the Si film 13 including the periphery of the gate electrode 17A is completely removed by isotropic dry etching using SF 6 gas. At this time, the gate electrode 17A made from Al is not etched by the SF 6 gas, and forms a Schottky junction with the operating layer 12.

Although this embodiment has exemplified the case using the Si film as an antireflection film, a TiN film and an SiC film can also be used. The etching gas in this case is SF 6 gas or CHF 3 gas.

FIGS. 3A to 3E are sectional views, respectively, for explaining a method of manufacturing a semiconductor device according to the second embodiment of the present invention.

With the same processing as in the first embodiment, after an operating layer 22 is formed on a semiconductor substrate 21, the surface of the operating layer 22 is selectively removed by wet etching or dry etching using a photoresist pattern as a mask to form a recess step 24, as shown in FIG. 3A. An amorphous carbon film (to be referred to as an α-C film hereinafter) 23 is sputtered on the operating layer 22, and a first positive photoresist film 25A is applied.

As shown in FIG. 3B, the obtained structure is exposed by an aligner (not shown) using a photomask (not shown). As an exposure beam, an i-line (λ=365 nm) or a KrF excimer beam (λ=248 nm) is used. In exposure, an exposure beam passing through the photomask is incident on an exposure region L in the first photoresist film 25A. Since the α-C film 23 below the first photoresist film 25A has a low transmittance with respect to the exposure beam, the exposure beam incident on the α-C film 23 is completely absorbed to be prevented from impinging on the first photoresist film 25A again. Since the α-C film 23 has a low reflectivity at the interface with the first photoresist film 25A, deformation of a pattern caused by reflection at the underlayer stage, waving of a pattern side wall due to the presence of a standing wave between incident and reflected beams, and variations in dimension caused upon changes in exposure energy by the multiple reflection effect in a resist film can be prevented. Subsequently, the resultant structure is developed by an alkali developing solution to dissolve the exposure region, forming a resist pattern having a resist opening 26A in the recess step 24.

As shown in FIG. 3C, the α-C film 23 is selectively removed by dry etching using O 2 gas and the remaining first photoresist film 25A as a mask, and a resist opening pattern is transferred to form an opening portion 26B. At the same time, the first photoresist film 25A is also etched and removed.

As shown in FIG. 3D, a metal film 27 is formed by, e.g., deposition on the entire surface including the opening portion 26B. At this time, the opening portion 26B is filled with the metal film 27. As this metal film, an Al film 27 or the like is used. Using the pattern of a second photoresist film 25B as a mask, the Al film 27 is selectively removed by dry etching using CCl 4 gas or BCl 4 gas to form a gate electrode 27A having a substantially T shape.

As shown in FIG. 3E, the second photoresist film 25B, and the α-C film 23 including the periphery of the gate electrode 27A are completely removed by isotropic dry etching using O 2 gas. At this time, the gate electrode 27A made from Al is not etched by the O 2 gas, and forms a Schottky junction with the operating layer 22. In this manner, according to the second embodiment, since the α-C film 23 and the second photoresist film 25B can be simultaneously etched, the process can be simplified more than the first embodiment.

Although the Al film is used as a gate electrode in the above embodiments, a WSi film or an Mo film can be used. The etching gas in this case is SF 6 gas.